The virtual screening of a database of molecules is based on explicit three-dimensional molecular superpositions. The torsional flexibility of the database molecules is taken fully into account, and an arbitrary number of conformation-dependent molecular features may be considered. A fragmentation-reassembly approach is utilized, which allows for an efficient sampling of the conformational space. A fast clique-based pattern-matching algorithm generates alignments of pairs of adjacent molecular fragments on the (rigid) query molecule that are subsequently reassembled to complete database molecules. Using conventional molecular features (hydrogen bond donors and acceptors, charges, and hydrophobic groups), it is possible to rapidly produce accurate alignments of medium-sized drug-like molecules. Examples with a test database containing a diverse set of 1780 drug-like molecules (including all conformers) show that average query processing times of the order of 0.1 seconds per molecule can be achieved on a PC, depending on the size of the query molecule.